Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,8 @@ model-index:
|
|
| 18 |
|
| 19 |
This model was cooked a bit too long during SIMPO training. It writes like Hunter S. Thompson 2 days into an ether binge. It's grotesque, dark, grimy and genius.
|
| 20 |
|
|
|
|
|
|
|
| 21 |
Strengths:
|
| 22 |
- Interesting, natural dialogue & character voices
|
| 23 |
- Surprising metaphors & unique turns of phrase
|
|
|
|
| 18 |
|
| 19 |
This model was cooked a bit too long during SIMPO training. It writes like Hunter S. Thompson 2 days into an ether binge. It's grotesque, dark, grimy and genius.
|
| 20 |
|
| 21 |
+
It's trained on an experimental [gutenberg + antislop dataset](sam-paech/gutenbergs_1_2_3_antislop-dpo). This contains the original two gutenberg sets by [jondurbin](jondurbin/gutenberg-dpo-v0.1) and [nbeerbower](nbeerbower/gutenberg2-dpo), as well as a subset of my own set, [gutenberg3](sam-paech/gutenberg3-generalfiction-scifi-fantasy-romance-adventure-dpo). The antislop pairs were generated with gemma-2-9b-it, with one sample generated with the [AntiSlop sampler](https://github.com/sam-paech/antislop-sampler) and the rejected sample generated without.
|
| 22 |
+
|
| 23 |
Strengths:
|
| 24 |
- Interesting, natural dialogue & character voices
|
| 25 |
- Surprising metaphors & unique turns of phrase
|